Plant Cell Physiol. 43(3): 256–265 (2002)
JSPP © 2002
Occurrence and Localization of Apocarotenoids in Arbuscular Mycorrhizal
Plant Roots
Thomas Fester 1, 4, Bettina Hause 1, Diana Schmidt 1, Kristine Halfmann 1, Jürgen Schmidt 1, Victor Wray 2,
Gerd Hause 3 and Dieter Strack 1
1
Institut für Pflanzenbiochemie, Abteilung Sekundärstoffwechsel, Weinberg 3, D-06120 Halle (Saale), Germany
Gesellschaft für Biotechnologische Forschung, Mascheroder Weg 1, D-38124 Braunschweig, Germany
3
Biozentrum der Universität Halle, Weinbergweg 22, D-06120 Halle (Saale), Germany
2
;
jected to dramatic changes as shown by the occurrence of
extensive plastid networks covering the arbuscules (Fester et
al. 2001). These networks are highly dynamic structures that
are formed and degraded concomitantly with the formation and
degradation of the arbuscules. They are probably responsible
for the biosynthesis of a variety of compounds (e.g. fatty acids
or nucleotides) important for the establishment and functioning of the symbiotic interface.
The biosynthesis of carotenoids in response to colonization by AM fungi is indicated by the accumulation of various
carotenoid degradation products (apocarotenoids) in AM roots.
Mycorradicin, an acyclic C14 polyene, i.e. 10,10¢-diapocarotene10,10¢-dioic acid, has first been described as the chromophore
of the yellow pigment in maize (Klingner et al. 1995a) and
wheat (Walter et al. 2000). The occurrence of a ‘yellow pigment’ in AM roots has been known for a long time (Jones
1924) and for a number of plants. In addition, C13 cyclohexenone derivatives, e.g. glycosylated 6-(3-hydroxybutyl)-1,1,5trimethyl-4-cyclohexen-3-ones (Maier et al. 1995), have been
observed in the AM roots from various members of the
Poaceae (Maier et al. 1997) and some Solanaceae (Maier et al.
1999, Maier et al. 2000). Mycorradicin and cyclohexenone
derivatives are regarded to be produced after the oxidative
cleavage of a precursor C40 carotenoid (Walter et al. 2000). A
protein catalyzing the corresponding cleavage reaction has
been described recently (Schwartz et al. 2001) for Arabidopsis
thaliana and Phaseolus vulgaris. Orthologs of the corresponding gene are present throughout the plant kingdom (Schwartz et
al. 2001).
In the present communication, we show that, along with
the water soluble C13 cyclohexenone derivatives mentioned
above, the yellow pigment serves for deposition of both apocarotenoids derived from the oxidative cleavage of a putative
precursor carotenoid molecule. According to our analysis, the
accumulation of mycorradicin in AM roots is much more widespread in the plant kingdom than previously thought. In addition, we provide evidence that the accumulation of the yellow
pigment occurs during the concomitant degradation of the fungal arbuscules and the covering plastid network.
The core structure of the yellow pigment from arbuscular mycorrhizal (AM) maize roots contains the apocarotenoids mycorradicin (an acyclic C14 polyene) and blumenol C cellobioside (a C13 cyclohexenone diglucoside). The
pigment seems to be a mixture of different esterification
products of these apocarotenoids. It is insoluble in water
and accumulates as hydrophobic droplets in the vacuoles of
root cortical cells. Screening 58 species from 36 different
plant families, we detected mycorradicin in mycorrhizal
roots of all Liliopsida analyzed and of a considerable
number of Rosopsida, but also species were found in which
mycorradicin was undetectable in mycorrhizal roots.
Kinetic experiments and microscopic analyses indicate that
accumulation of the yellow pigment is correlated with the
concomitant degradation of arbuscules and the extensive
plastid network covering these haustorium-like fungal
structures. The role of the apocarotenoids in mycorrhizal
roots is still unknown. The potential C40 carotenoid precursors, however, are more likely to be of functional importance in the development and functioning of arbuscules.
Key words: Apocarotenoids — Arbuscular mycorrhiza —
Cyclohexenone derivatives — Mycorradicin — Yellow pigment — Zea mays.
Abbreviations: AM, arbuscular mycorrhiza; MS, mass spectrometry; ES, electrospray; NMR, nuclear magnetic resonance; UV-Vis,
ultraviolet-visible.
Introduction
The arbuscular mycorrhiza (AM) is a mutualistic interaction between fungal species from the order Glomales (Zygomycetes) and roots of most terrestrial plants (for review see
Smith and Read 1997). The key feature of this symbiosis is the
arbuscule, a highly branched haustorium-like fungal structure
within root cortical cells, that constitutes the symbiotic interface of nutrient exchange.
During development of arbuscules, root plastids are sub4
Corresponding author: E-mail, [email protected]; Fax, +49-345-5582-1009; Phone, +49-345-5582-1521.
256
Apocarotenoids in arbuscular mycorrhizal roots
Results
Hydrolytic fragments of the yellow pigment
The yellow pigment was soluble in polar organic solvents
(e.g. methanol or dioxan), but insoluble in water as well as in
less polar organic solvents (e.g. n-butanol, ethyl acetate,
dichloromethane). After purification by subsequent TLC and
HPLC the pigment eluted as a broad, bell-shaped signal from
analytical HPLC (Fig. 1A). In the course of the purification
procedure all cyclohexenone derivatives described by Walter et
al. (2000) were removed due to the differing solubilities in
water and their different retention times on HPLC.
Addition of KOH (500 mM final concentration in 80% aq.
methanol) and incubation for 1 h at room temperature led to
hydrolysis of the yellow pigment yielding mycorradicin (Fig.
1B, peak 2b) and several other compounds. Mycorradicin was
identified by ES-MS with analogous results (not shown) to the
data published by Klingner et al. (1995a) for mycorradicin
dimethyl ester. Irradiation of purified mycorradicin with day
light resulted in splitting of the signal into four peaks of different intensities (Fig. 1B, peaks 2a, 2b, 2c and 2d), which was
probably due to the formation of cis–trans isomers. We always
extracted a mixture of these isomers, unless harvesting, extraction and hydrolysis were conducted under red safety light conditions, when only one product (Fig. 1B, peak 2b) was found
(data not shown). For practical reasons it was not possible to
apply these safety light conditions during routine extraction or
preparative work.
Another prominent compound of the hydrolysate was
identified as blumenol C cellobioside (Fig. 1B, peak 1a). The
positive ion ES mass spectrum of this compound obtained by
LC-MS was similar to those of blumenin and nicoblumin
(Maier et al. 1995, Maier et al. 1999). The structure of the compound was readily established from the 1D and 2D NMR spectra and comparison with data from related compounds (Maier
et al. 2000). Unlike the disaccharide units found previously
(Maier et al. 1995, Peipp et al. 1997, Maier et al. 1999, Maier
et al. 2000), the present compound did not possess a 1²-6¢-, but
a 1²-4¢-sugar linkage. Thus the compound is 6-(9-hydroxybutyl)1,1,5-trimethyl-4-cyclohexen-3-one 9-O-b-glucopyranosyl-(1,4)b-glucopyranoside, i.e. blumenol C 9-O-b-cellobioside. Apart
from mycorradicin and blumenol C cellobioside, a minor, unidentified compound absorbing at 309 nm (peak 3 in Fig. 1B)
was liberated by alkaline treatment of the yellow pigment.
257
Mycorradicin in mycorrhizal roots from different plant families
Fifty-eight species of 36 different plant families were
analyzed for the occurrence of mycorradicin in mycorrhizal
roots. As summarized in Table 1, we detected the compound
in the mycorrhizal roots from all Liliopsida examined as well
as in mycorrhizal roots from a number of Rosopsida belonging
to the Cucurbitaceae and Fabaceae (subclass Rosidae) as well
as to the Polemoniaceae, Solanaceae, Scrophulariaceae, and
Apiaceae (subclass Asteridae). In contrast, we found many
other species from the Rosopsida which were strongly colonized, but did not accumulate mycorradicin (e.g. Ruta graveolens, Hypericum perforatum, Linum usitatissimum, Petroselinum crispum, Apium graveolens).
In cases where mycorradicin occurred in mycorrhizal
roots, its amount appeared species-specifically and correlated
clearly with the yellow coloration of the roots. To some degree,
the amount of mycorradicin was dependent on the fungal partner as well, as can be seen for Medicago truncatula and Zea
mays. Both plants produced higher amounts of mycorradicin
after inoculation with Glomus intraradices when compared to
inoculation with Glomus mosseae. This may be partly due to
more efficient colonization by G. intraradices.
In few cases (Z. mays dwarf 1-mutants, Nardus stricta,
Cucumis sativus, Cucurbita pepo, P. vulgaris, Ornithogalum
umbellatum), mycorradicin could be found in non-mycorrhizal
roots, but always to a much lower extent compared to the
mycorrhizal roots (Table 1). In maize this phenomenon was
observed only for the dwarf 1-mutant but not for the wild-type
plants (data not shown). In all the other plants examined, the
accumulation of the compound was unique to mycorrhizal
roots. Moreover, a further increase in accumulation could not
be induced by treatment of non-mycorrhizal maize (dwarf 1mutant) plants with elevated amounts of mineral nutrients
(nitrate, phosphate) or hormones (kinetin, methyl jasmonate,
indole-3-acetic acid, gibberellin A3). All treated plants contained mycorradicin in the range of 0.7–5.9 nmol (g FW)–1,
comparable to mycorradicin contents of untreated control plants.
Localization of the yellow pigment
Using the nurse culture system, we observed mycorrhizal
colonization of wheat and maize (dwarf 1-mutant) roots 8 d
after transplanting the plants. First root segments showing yellow coloration were visible after 16 d. A significant increase in
mycorradicin was observed after 24 d for maize (not shown)
and after 20 d for wheat (Fig. 2).
Analysis of the maize roots by epifluorescence micros-
Fig. 1 Analytical HPLC-chromatogram (identical elution conditions; eluent: acetonitrile/water/phosphoric acid in varying proportions) of nontreated (A) and alkaline-treated (B) yellow pigment. Insets show UV-Vis-spectra of the main components. (A) The yellow pigment elutes as a
broad bell-shaped peak from analytical HPLC (absorption maxima at 240, 318 and 382 nm). (B) After adjusting to 500 mM KOH, hydrolytic
fragments can be separated by HPLC. Compound 1a and 1b are two isomeric forms of blumenol C cellobioside (absorption maxima at 245 nm).
Compounds 2a, 2b, 2c and 2d are isomeric forms of mycorradicin, whose appearance is induced by light. Compound 2b is the isomeric form that
is obtained under red safety light conditions. 2b, 2c and 2d show the UV-Vis spectrum depicted in the inset, 2a has a slightly different spectrum
(absorption maximum at 375 nm). Compound 3 is also liberated from the yellow pigment and has an absorption maximum of 309 nm.
258
Apocarotenoids in arbuscular mycorrhizal roots
Fig. 1
Apocarotenoids in arbuscular mycorrhizal roots
259
Table 1 Occurrence of mycorradicin in mycorrhizal roots (microsymbiont Glomus intraradices, if not indicated otherwise) from
various plant families
Taxon
Pteridopsida
Adiantaceae
Adiantum sp.
Aspidiaceae
Polystichum lonchitis L.
Davalliaceae
Nephrolepsis exaltata L.
Magnoliopsida
Aristolochiaceae
Asarum europaeum L.
Magnoliaceae
Magnolia grandiflora L.
Rosopsida
Amaranthaceae
Amaranthus paniculatus L.
Apiaceae
Anethum graveolens L.
Apium graveolens L.
Daucus carota L.
Foeniculum vulgare Miller
Petroselinum crispum (Miller) A.W. Hill
Boraginaceae
Myosotis arvensis L.
Calendulae
Calendula officinalis L.
Campanulaceae
Campanula sp.
Cannabaceae
Cannabis sativa L.
Cardueae
Centaurea cyanus L.
Cucurbitaceae
Cucurbita pepo L.
Cucumis sativus L.
Fabaceae
Lotus japonicus (Regel) K. Larsen
Medicago sativa L.
Lupinus polyphyllus Lindl.
Medicago truncatula Gartn./Glomus
intraradices Schenck & Smith
Medicago truncatula Gaertn./Glomus
mosseae Gerd. & Trappe
Phaseolus vulgaris L
Trifolium arvense L.
Geraniaceae
Geranium robertianum L.
Pelargonium sp.
Guttiferae
Hypericum perforatum L.
Helenieae
Tagetes erecta L.
Heliantheae
Helianthus annuus L.
a
Mycorrhiza a Mycorradicin b Taxon
–
n.d.
–
n.d.
–
n.d.
+
n.d.
++
n.d.
–
n.d.
+++
+++
+
+++
+++
0.8
n.d.
n.d.
0.4
n.d.
+
n.d.
–
n.d.
+
n.d.
++
n.d.
+++
n.d.
++
++
14 (2.6) c
96 (2.0) c
+++
++
–
+++
1.9
17
n.d.
17
++
7.2
++
++
0.8 (tr.) c
31
–
+++
n.d.
tr.
+++
n.d.
+++
n.d.
+++
n.d.
Lactucaceae
Cichorium intybus L.
Lactuca sativa L.
Linaceae
Linum usitatissimum L.
Lobeliaceae
Lobelia erinus L.
Malvaceae
Malva sp.
Onagraceae
Oenothera missuriensis Sims.
Papaveraceae
Papaver rhoeas L.
Papaver somniferum L.
Polemoniaceae
Phlox drummondii Hook.
Primulaceae
Primula auricula L.
Ranunculaceae
Aquilegia vulgaris L.
Delphinium ajacis L.
Rutaceae
Ruta graveolens L.
Salicaceae
Populus tremula x alba
Scrophulariaceae
Antirrhinum majus L.
Digitalis purpurea L.
Solanaceae
Lycopersicon esculentum Miller
Nicotiana tabacum L.
Solanum tuberosum L.
Valerianaceae
Valerianella locusta L.
Violaceae
Viola x wittrockiana Gams
Liliopsida
Alliaceae
Allium cepa L.
Allium porrum L.
Mycorrhiza a Mycorradicin b
++
+++
n.d.
n.d.
+++
n.d.
+
n.d.
–
n.d.
–
n.d.
+++
–
n.d.
n.d.
+
0.5
++
n.d.
+++
+
tr.
n.d.
+++
n.d.
–
n.d.
++
+++
n.d.
tr.
++
+++
+++
0.9
16
28
++
n.d.
–
n.d.
+++
+++
5.5
1
Liliaceae
Asparagus densiflorus (Kunth) J.P. Jessop
Ornithogalum umbellatum L.
Poaceae
Nardus stricta L.
Panicum miliaceum L. convar. Miliaceum var. miliaceum
Sorghum bicolor (L.)
Moench S.L.
Zea mays L. dwarf 1-mutant/Glomus
intraradices Schenck & Smith
Zea mays L. dwarf 1-mutant/Glomus
mosseae Gerd. & Trappe
++
++
tr.
130 (0.3) c
+++
+++
4 (0.5) c
15
+++
4
+++
310 (2.8) c
+++
220 (2.8) c
Mycorrhization was estimated after staining with trypan blue; +, ++, +++ = less than 30, between 30 and 60, more than 60% mycorrhizal colonization, respectively.
b
Mycorradicin values are given as nmol (g FW)–1 referring to crocetin, the C20 homolog of mycorradicin as external standard.
c
Mycorradicin in non-mycorrhizal plants (values given in brackets).
– = Not colonized; n.d. = not detectable; tr. = trace amounts.
260
Apocarotenoids in arbuscular mycorrhizal roots
copy supported the data obtained by chemical analysis. In nonmycorrhizal roots, only a few single cortical cells contained
small yellow fluorescent droplets (Fig. 3A). This correlated
with the low amounts of mycorradicin extracted from these
roots as already mentioned. The cortical cells harboring the
first generation of arbuscules also contained little or no yellow
fluorescent droplets (Fig. 3B). Such droplets were sometimes
visible in the cytoplasmic seam surrounding the vacuole (not
shown). The yellow root segments formed about 1 week after
the initial colonization (i.e. after 2 weeks cultivation) always
showed large amounts of the yellow fluorescent material,
which in some cases seemed to fill up the complete cellular
space (Fig. 3C). The yellow fluorescent droplets could be identified to contain the yellow pigment by microspectral photometry. They exhibited an absorption maximum at 390 nm with
shoulders at 375 and 414 nm. This corresponds to an UV-Vis
spectrum of the extracted yellow pigment with an absorption
maximum at 382 nm and two shoulders at 365 and 405 nm,
regarding a bathochromic shift of 10 nm (Fig. 3D). Spectra
were recorded above 330 nm since high background absorption occurred below this wavelength.
Fig. 2 Mycorrhizal structures and mycorradicin contents in wheat
roots after transplantation to a nurse-culture-system. Roots were harvested after transplantation as indicated. First traces of mycorradicin
were measured only 16 d after the development of mycorrhizal structures.
Fig. 3 Epifluorescence micrographs of non-mycorrhizal and mycorrhizal maize roots. (A) Non-mycorrhizal roots: yellow fluorescing droplets
can be observed in single cortical cells. (B) Mycorrhizal root after two weeks of inoculation: arbuscules show only slight fluorescence different
from the yellow pigment. (C) Micrograph of a yellow root segment from a mycorrhizal plant after three weeks of inoculation: yellow colored root
segments always show mycorrhizal colonization and always contain large amounts of yellow fluorescing droplets. In some cases these droplets
are fused forming larger structures. (D) Comparison between the UV-Vis spectra of purified yellow pigment (1) and the intracellular yellow droplets (2). Bars represent 100 mm in A, B and C.
Apocarotenoids in arbuscular mycorrhizal roots
261
Fig. 4 Transmission electron micrographs of ultra thin sections of mycorrhizal root cortical cells of maize dwarf-1 plants. (A) Overview of a cell
containing intact arbuscular structures located in the close neighborhood of plant cell organelles. Note the presence of only few osmiophilic droplets; a, arbuscule; n, plant nucleus; v, plant vacuole; bar represents 1 mm. (B) Detail of a cell containing collapsing arbuscules (arrows). Near the
arbuscular structures plant organelles and cytoplasm as well as large osmiophilic droplets are visible; pt, plastid; bar represents 0.5 mm. (C) Overview of a cell harboring a collapsing arbuscule (arrow). Note the high accumulation of osmiophilic droplets within the cytoplasm; bar represents
1 mm. (D) Detail of a cell containing a collapsed arbuscule; osmiophilic droplets are located within the cytoplasm; m, mitochondrion; arrow,
endoplasmic reticulum; bar represents 0.25 mm. (E) Overview of a cell harboring a nearly fully collapsed arbuscule (arrows). The osmiophilic
droplets are accumulated within the vacuole; a, intact arbuscular structure; v, plant vacuole; bar represents 2.5 mm.
Electron microscopic analysis of mycorrhizal roots from
the maize dwarf 1-mutant revealed an accumulation of small
osmiophilic droplets in cells containing fungal structures
(Fig. 4). These droplets were not observed in roots from nonmycorrhizal roots (not shown). They were found mainly close
to collapsing arbuscules (Fig. 4C, E), but not in the surrounding of intact arbuscules (Fig. 4A). Their location within the cell
correlated with the number of droplets present: in cells contain-
ing only few droplets, they were located inside the cytoplasm
(Fig. 4B–D); in those containing high numbers of droplets they
accumulated within the vacuole (Fig. 4E).
Discussion
Mycorradicin and C13 cyclohexenone derivatives are
regarded to be produced after oxidative cleavage of one precur-
262
Apocarotenoids in arbuscular mycorrhizal roots
Fig. 5 Hypothetical scheme of the yellow pigment biosynthesis starting from a putative precursor carotenoid, e.g. zeaxanthin (A): (1) Carotenoid dioxygenase activity: oxidative cleavage leading to two C13 ketones that are further reduced to blumenol C (B) and a C14 dialdehyde (see
Baskin 1992) that is further oxidized to mycorradicin (C). (2) Glucosyltransferase activities: glucosylation of blumenol C leading to blumenol C
cellobioside (D). (3) Mycorradicin transferase activity: ester formation of mycorradicin with the glucosyl residues of blumenol C cellobioside (E).
The product (E) may enter further esterification reactions with mycorradicin adding further blumenol C cellobiosides to give a highly complex
network of polyester structures.
sor carotenoid molecule at positions 9,10 and 9¢,10¢ (Walter et
al. 2000). In previous reports, the occurrence of mycorradicin
has been shown only for AM roots from maize and wheat
(Klingner et al. 1995a, Walter et al. 2000), whereas cyclohexenone derivatives have been observed in the AM roots from various members of the Poaceae (Maier et al. 1997) and of some
Solanaceae (Maier et al. 1999, Maier et al. 2000). Here we
show that mycorradicin accumulates in AM roots of a large
number of plants and that in AM maize roots, along with some
water soluble cyclohexenone derivatives, both carotenoid degradation products are deposited in the same molecular form, i.e.
as the yellow pigment.
The various cyclohexenone derivatives described by
Maier et al. (1995), Maier et al. (1997), Maier et al. (1999),
Maier et al. (2000) can be detected as water-soluble components besides the yellow pigment (Walter et al. 2000), indicating that only a part of the C13 carotenoid cleavage products
is integrated into the yellow pigment. Due to different chemical properties, they can be separated very easily from the yellow pigment and did not interfere with our analysis. In contrast
to the C13 carotenoid cleavage products, we were not able to
detect free mycorradicin, not incorporated into the yellow pigment complex. The different result regarding this point presented by Klingner et al. (1995a) might be explained by our
observation that even 1 mM KOH (final concentration) is sufficient to liberate the dimethyl ester of mycorradicin from the
Apocarotenoids in arbuscular mycorrhizal roots
yellow pigment (data not shown).
All our attempts to elucidate the complete structure of the
yellow pigment failed. From the results obtained, however, we
assume that the core structure of the yellow pigment consists of
a complex mixture of polyesters of mycorradicin and glycosylated cyclohexenone derivatives, i.e. blumenol C cellobioside in maize roots. This assumption leads to a hypothetical
scheme for the biogenesis of the yellow pigment (Fig. 5),
where yet unknown precursor carotenoids (Fig. 5A) are oxidatively cleaved. The resulting C14 dialdehyde (see Schwartz et
al. 2001) is then oxidized to mycorradicin (Fig. 5C), whereas
the resulting C13 ketones are reduced to blumenol C (Fig. 5B)
and glucosylated to blumenol C cellobioside (Fig. 5D). Finally,
the sugar moieties of this compound are esterified with mycorradicin (Fig. 5E, position unknown). A well known example of
a similar structure is the digentiobiose ester of crocetin (acrocin), the main pigment of saffron (Pfander and Wittwer
1975). Crocetin is the C20 homolog of mycorradicin from Gardenia sp. and Crocus sp.
Maier et al. (1997) had described the accumulation of
cyclohexenone derivatives in the mycorrhizal roots of a large
number of Poaceae (Liliopsida). Due to the low detection limit
of mycorradicin in the HPLC analysis, we could detect this
compound even in those species which had been described to
be devoid of cyclohexenone derivatives, namely Panicum miliaceum, Nardus stricta and Sorghum bicolor. Whereas the
accumulation of apocarotenoids seems to be a common characteristic of Liliopsida species, we failed to detect mycorradicin
in AM roots of a number of Rosopsida species (Table 1).
Regarding this class, there was no correlation between the phylogenetic position of a given plant and the accumulation of
mycorradicin in AM roots.
In some plant species (Z. mays dwarf 1-mutants, Nardus
stricta, C. sativus, C. pepo and P. vulgaris), we found small
amounts of mycorradicin in non-mycorrhizal roots. In all these
cases, mycorradicin contents of the corresponding mycorrhizal
roots were markedly higher (Table 1). When we tried to
increase artificially the basal level of mycorradicin present in
non-mycorrhizal maize (dwarf 1-mutant) roots, neither supply
of mineral nutrients, nor hormone treatments induced mycorradicin accumulation. The maize dwarf 1-mutants (deficient in
gibberellin biosynthesis) were used due to their rapid accumulation of the yellow pigment upon fungal colonization. In addition, since these mutants show a basic level of mycorradicin in
non-mycorrhizal roots, we assumed that they would react more
sensitively to the treatments with hormones and mineral nutrients. Our data regarding this point support earlier experiments
by Klingner et al. (1995b) and Maier et al. (1997) clearly showing that apocarotenoid production in mycorrhizal roots, apart
from a basal level in the rare cases mentioned, is closely correlated to mycorrhization.
There are several lines of evidence indicating that the
accumulation of the yellow pigment takes place during the disintegration of the arbuscules. Kinetic experiments showed that
263
accumulation of the yellow droplets as well as increased
amounts of extractable mycorradicin were not observed at the
early stages of the root mycorrhization. Roots with first mycorrhizal structures did not contain either increased amounts of
mycorradicin or any significant formation of yellow droplets.
One week later, yellow colored mycorrhizal root segments
could already be seen with the naked eye. In agreement with
Klingner et al. (1995b), who described massive deposition of
the yellow pigment in vacuoles of some mycorrhizal gramineous plants, we observed massive accumulation of yellow material, often filling up the entire cell in these root segments.
According to data from microspectral photometry, mycorradicin was responsible for the yellow coloration of this material. The time period between the first observation of arbuscules and the beginning accumulation of mycorradicin
corresponded roughly with the life span reported for arbuscules (Smith and Read 1997). Electron microscopy did not
allow a definitive localization of the yellow pigment, but provided additional evidence for a connection of the degradation
of arbuscules and accumulation of the yellow pigment. We
observed large numbers of hydrophobic droplets specifically in
mycorrhizal root sections and particularly abundant close to
disintegrating arbuscular structures. Taken together, our results
indicate that the apocarotenoids constituting the yellow pigment are produced during the degradation of arbuscules and in
particular during the disintegration of the plastid network covering these arbuscules (Fester et al. 2001).
Referring to possible biological functions of these apocarotenoids, there is only one report showing a negative influence
of blumenin on the development of AM (Fester et al. 1999). In
contrast to this, cyclohexenone derivatives were not responsible for the systemic suppression of mycorrhization in precolonized barley plants (Vierheilig et al. 2000), and mycorradicin
did not suppress an elicitor-induced oxidative burst reaction in
Nicotiana tabacum and Medicago sativa cell cultures (Schröder
et al. 2001). Furthermore, a direct functional role seems
unlikely, because the yellow pigment is insoluble in water and
because of its site of deposition described above. However, during the years of investigating apocarotenoids from barley,
maize and wheat roots, we never observed mycorrhizal colonization of these plants without concomitant accumulation of
apocarotenoids. Studies have been initiated to elucidate the
structure of the precursor carotenoids of the yellow pigment in
these plants and to evaluate their possible role in the formation
and maintenance of functional arbuscular mycorrhizas.
Materials and Methods
Plant material and AM-fungus inoculation
Geranium robertianum seeds were from Bornträger GmbH, (Offstein, Germany) and Ruta graveolens, Panicum miliaceum, Nardus
stricta and Sorghum bicolor seeds from the botanical garden, Halle.
Papaver somniferum, Papaver rhoeas and Cannabis sativa seeds were
a kind gift from Prof. Kutchan (Institute of Plant Biochemistry, Halle,
Germany). All the other plants or seeds were from N.L. Chrestensen
264
Apocarotenoids in arbuscular mycorrhizal roots
(Erfurt, Germany), except for Tagetes erecta, Z. mays (dwarf 1mutant), Delphinium ajacis and N. tabacum which were available in
our institute garden. Plants were grown in 250-ml plastic pots filled
with expanded clay (Lecaton, 2–5 mm particle size, Fibo Exclay, Pinneberg, Germany) and inoculated with the AM fungus Glomus intraradices Schenck & Smith (isolate 49, propagules in expanded clay
provided by H. von Alten from the collection of the Institut für Pflanzenkrankheiten und Pflanzenschutz der Universität Hannover, Germany) or Glomus mosseae (BEG12, propagules in calcinated clay provided by Société Biorize, Dijon, France).
Formation of mycorrhizas was induced by growing the plants for
at least 6 weeks in expanded clay mixed with 10% (v/v) of the fungal
inoculum. Details of plant growth conditions have been published previously (Maier et al. 1995). The approximate percentage values for
mycorrhiza formation were estimated microscopically after staining
with trypan blue in lactophenol according to a procedure described by
Phillips and Hayman (1970).
Kinetic experiments
To ensure fast and synchronous mycorrhization of the roots, Z.
mays (dwarf 1-mutants) and Triticum aestivum plants were grown in a
nurse-culture system with leek as nurse plant (Rosewarne et al. 1997).
Mycorrhizal leek roots were grown by cultivating four leek plants per
pot for 6 weeks together with G. intraradices inoculum. After this
time, young maize or wheat plants were transplanted into these pots.
For kinetic experiments, five plants were harvested every 4 d and used
to determine the degree of mycorrhization and accumulation of mycorradicin. The percentage of root length colonized by G. intraradices
was determined microscopically with the gridline-intersection method
at a magnification of ´20.
Treatment of non-mycorrhizal plants with phytohormones and mineral
nutrients
Zea mays dwarf 1-mutants were cultivated without fungal inoculum for 6 weeks. Then plants were watered with 10 ml of either kinetin
(4.6 mM), methyl jasmonate (45 mM), indoleacetic acid (5 mM) or
gibberellin A3 (2.6 mM) daily for 1 week. Four-week-old maize plants
were watered daily for 1 week with 10 ml of a solution containing a
10-fold concentration of either nitrate or phosphate (400 mM nitrate,
30 mM phosphate) compared to the control. The plants were harvested and extracted as described below.
Extraction and quantification of mycorradicin
Freshly harvested roots were washed with water and 2 g root tissue were ground in liquid nitrogen and extracted once for 30 min with
4 ml 80% aq. methanol. The mixture was centrifuged and the supernatant was adjusted to 0.5 M KOH incubated for 1 h at room temperature and centrifuged again before being subjected to analytical HPLC.
The liquid chromatograph (600-MS system controller, Waters, Milford, U.S.A.) was equipped with a 5-mm Nucleosil C18 column
(250´4 mm i.d.; Macherey-Nagel, Düren, Germany), and a linear gradient elution system was applied at a flow rate of 1 ml min–1 within
35 min from solvent A (1.5% ortho-phosphoric acid in water) to solvent B (100% acetonitrile). Injections of 20 ml were carried out with an
automatic sampler (717 autosampler, Waters, Milford, U.S.A.). Compounds were detected photometrically (maxplot between 200 and
500 nm) with a Waters (Milford, U.S.A.) 996 photodiode array detector. Quantitative values were calculated from external standardization
with crocetin (Sigma, Deisenhofen, Germany) using the Millenium
software 2010 (Millipore, Eschborn, Germany).
Isolation of the yellow pigment
Methanolic extracts (80% aq. methanol) from the mycorrhizal
maize roots (dwarf 1-mutant, about 20 g FW) were filtered and evaporated at 40°C (in vacuo) until close to dryness. After addition of water,
the suspension was filtered or centrifuged. The insoluble material was
redissolved in 80% aq. methanol. Further purification was performed
by thin layer chromatography (TLC) using four plates coated with silica gel 60 (Merck, Darmstadt, Germany) and the upper phase of
n-butanol : acetic acid : water (5 : 1 : 4, by vol.) as a solvent. After TLC
for about 10 h, the yellow band was scraped off and the material
extracted with ethyl acetate and n-butanol. Finally the yellow pigment, extracted from the silica gel with 80% aq. methanol, was further
purified by HPLC. The liquid chromatograph (System Gold; Beckman Instruments, München, Germany) was equipped with a Nucleosil
100–10 C18 column (VarioPrep; 10 mm, 250´40 mm i.d.; MachereyNagel, Düren, Germany). Compounds were eluted using a linear gradient from solvent A (1% aq. acetic acid, 30% methanol) to solvent B
(90% aq. methanol) at a flow rate of 10 ml min–1 within 70 min and
further 40 min with solvent B. The fractions containing the yellow pigment were collected, evaporated at 40°C (in vacuo) to dryness and further analyzed by analytical HPLC.
Isolation of hydrolytic fragments from the yellow pigment
The purified yellow pigment was dissolved in 80% aq. methanol, the solution was adjusted to 0.5 M KOH (final concentration) and
incubated for 1 h at room temperature. The alkaline solution was
centrifuged and fractionated by HPLC using the chromatograph mentioned above and application of a linear gradient elution system at a
flow rate of 10 ml min–1 within 80 min from solvent A (1% aq. acetic
acid) to solvent B (90% aq. methanol) and a further 40 min with
solvent B. Further purification of the cyclohexenone derivative was
achieved by HPLC using the Waters (Milford, U.S.A.) 600-MS system
controller, a 5-mm Nucleosil C18 column (250´4 mm i.d.; MachereyNagel), and a linear gradient elution system at a flow rate of 1 ml min–1
within 35 min from solvent A (0.5% formic acid in water, 23%
acetonitrile) to solvent B (0.5% formic acid in water, 27%
acetonitrile). Quantitative values for cyclohexenone derivatives were
calculated from external standardization with ABA (Fluka, Buchs,
Germany) according to Maier et al. (1995).
Mass spectrometry (MS)
The electrospray (ES) mass spectra were obtained from a Finnigan (San José, U.S.A.) MAT TSQ 7000 instrument (electrospray voltage 4.5 kV (positive ions), 3.5 kV (negative ions); heated capillary
(220°C; sheath gas nitrogen) coupled with a Micro-Tech Ultra-Plus
MicroLC system equipped with a 4-mm ULTRASEP C18 column
(100´1 mm i.d.). For HPLC a linear gradient elution system at a flow
rate of 70 ml min–1 within 15 min from solvent A (0.2% acetic acid in
water, 15% acetonitrile) to solvent B (0.2% acetic acid in water, 90%
acetonitrile) was chosen. Blumenol C cellobioside was detected at Rt =
13.6 min, positive ion ES-MS m/z (rel. int.): 535 ([M + H]+, 92), 373
([M + H – C6H10O5]+, 28), 211 ([M + H – 2 x C6H10O5]+, 92); negative
ion ES-MS m/z (rel. int.): 533 ([M – H]–, 100).
Nuclear magnetic resonance (NMR) spectroscopy
1D (1H) and 2D (COSY and HMBC) NMR spectra of the
cyclohexenone derivative isolated from the purified yellow pigment
after alkaline treatment were recorded at 27°C on a Bruker AVANCE
DMX 600 NMR spectrometer (Karlsruhe, Germany) locked to the
major resonance of the solvent, CD3OD. Chemical shifts are in ppm
and coupling constants in Hz. The compound numbering scheme is
that used in Maier et al. (2000): d = 5.84 [bs, H-4], 4.45 [d, H-1², J(1²2²) 7.9], 4.39 [d, H-1¢, J(1¢-2¢) 7.8], 3.93–3.89 [m, H-6¢A/B, H-6²A],
3.85 [m, H-9], 3.70 [dd, H-6²B, J(6²A-6²B) 11.9, J(6²B-5²) 5.7], 3.59
[dd, H-4¢, J(4¢-3¢) 9.0, J(4¢-5¢) 9.1], 3.53 [dd, H-3¢, J(2¢-3¢) 8.9], 3.44–
Apocarotenoids in arbuscular mycorrhizal roots
3.32 [m, H-5¢, H-3², H-4², H-5²], 3.26 [dd, H-2², J(2²-3²) 9.0], 3.25
[dd, H-2¢], 2.52 [d, H-2A, J(2A-2B) 17.4], 2.09 [bs, H-13], 2.02 [d, H2B], 2.02 [m, H-6], 1.85 [m, H-7A], 1.76–1.62 [m, H-7B, H-8A/B],
1.29 [d, H-10, J(10–9) 6.3], 1.14 [s, H-11], 1.06 [s, H-12]. The 13C
chemical shifts were taken from a complete set of correlations in the
2D HMBC spectrum: d = 202.0 (s, C-3), 169.5 (s, C-5), 125.1 (d, C-4),
104.2 (d, C-1²), 103.6 (d, C-1¢), 80.4 (d, C-4¢), 77.5 (d, C-9), 77.5 (3 x
d, C-5¢, C-3², C-5²), 76.2 (d, C-3¢), 74.6 (2 x d, C-2¢, C-2²), 71.0 (d, C4²), 62.1 (t, C-6²), 61.7 (t, C-6¢), 52.4 (d, C-6), 47.8 (t, C-2), 37.2 (s,
C-1), 37.2 (t, C-8), 28.7 (q, C-12), 27.1 (q, C-11), 26.4 (t, C-7), 24.6
(q, C-13), 21.6 (q, C-10).
Microscopical methods
Using a Nikon (Tokyo, Japan) SMZ-U stereomicroscope and thin
needles the freshly harvested roots of non-mycorrhizal and mycorrhizal maize plants (dwarf 1-mutant) were disrupted on glass slides in
50 mM HEPES (pH 7.0), 125 mM sorbitol, 2.5 mM b-mercaptoethanol,
2.5 mM EDTA in order to generate thin preparations of the root cortex.
Bright field and epifluorescence microscopy was performed using a
Nikon Optiphot-2 with the following filter settings for epifluorescence:
EX330~380, DM400, BA435. Photographs were taken using Fujifilm
(Tokyo, Japan) Sensia II. The slides were scanned and processed using
the programs PhotoPaint 7.0 (Corel Corporation, Ottawa, Canada).
Spectra from intracellular droplets were measured using the above
mentioned preparations and a MPM 800 microspectral photometer
from Zeiss (Jena, Germany). They were recorded from 330 to 500 nm,
with a 2.5 nm optical bandwidth as quotient spectra comparing a region
from a large yellow droplet within a cortical cell and a region of the
same size in a similar cortical cell without the yellow material.
Electron microscopy
Small pieces of the mycorrhizal roots were dissected and fixed in
3% glutardialdehyde in 0.1 M sodium cacodylate buffer (pH 7.0) for
3 h at room temperature. After rinsing with buffer, the samples were
postfixed with 1% osmium tetroxide in the buffer for 30 min, rinsed
again and dehydrated in a graded ethanol series. Ethanol was substituted by butylmethyl methacrylate at 4°C and samples were polymerised by UV light at 4°C for 24 h (Baskin et al. 1992). Ultra thin sections were poststained with uranyl acetate and lead citrate. Sections
were visualized with a Zeiss TEM 900 electron microscope.
Acknowledgments
We thank Gerlinde Waiblinger for skillful technical assistance.
This work was supported by the Deutsche Forschungsgemeinschaft
and the Fonds der Chemischen Industrie.
References
Baskin, T.I., Busby, C.H., Fowke, L.C., Sammut, M. and Gubler, F. (1992)
Improvements in immunostaining samples embedded in methacrylate: Locali-
265
zation of microtubules and other antigens throughout developing organs in
plants of diverse taxa. Planta 187: 405–413.
Fester, T., Maier, W. and Strack, D. (1999) Accumulation of secondary compounds in barley and wheat roots in response to inoculation with an arbuscular mycorrhizal fungus and co-inoculation with rhizosphere bacteria.
Mycorrhiza 8: 241–246.
Fester, T., Strack, D. and Hause, B. (2001) Reorganization of tobacco root plastids during arbuscule development. Planta 213: 864–868.
Jones, F.R. (1924) A mycorrhizal fungus in the roots of legumes and some other
plants. J. Agric. Res. 29: 459–470.
Klingner, A., Bothe, H., Wray, V. and Marner, F.J. (1995a) Identification of a
yellow pigment formed in maize roots upon mycorrhizal colonization. Phytochemistry 38: 53–55.
Klingner, A., Hundeshagen, B., Kernebeck, H. and Bothe, H. (1995b) Localization of the yellow pigment formed in roots of gramineous plants colonized by
arbuscular fungi. Protoplasma 185: 50–57.
Maier, W., Hammer, K., Dammann, U., Schulz, B. and Strack, D. (1997) Accumulation of sesquiterpenoid cyclohexenone derivatives induced by an arbuscular mycorrhizal fungus in members of the Poaceae. Planta 202: 36–42.
Maier, W., Peipp, H., Schmidt, J., Wray, V. and Strack, D. (1995) Levels of a
terpenoid glycoside (blumenin) and cell wall-bound phenolics in some cereal
mycorrhizas. Plant Physiol. 109: 465–470.
Maier, W., Schmidt, J., Nimtz, M., Wray, V. and Strack, D. (2000) Secondary
products in mycorrhizal roots of tobacco and tomato. Phytochemistry 54:
473–479.
Maier, W., Schmidt, J., Wray, V., Walter, M.H. and Strack, D. (1999) The arbuscular mycorrhizal fungus, Glomus intraradices, induces the accumulation of
cyclohexenone derivatives in tobacco roots. Planta 207: 620–623.
Peipp, H., Maier, W., Schmidt, J., Wray, V. and Strack, D. (1997) Arbuscular
mycorrhizal fungus-induced changes in the accumulation of secondary compounds in barley roots. Phytochemistry 44: 581–587.
Pfander, H. and Wittwer, F. (1975) Untersuchungen zur Carotinoid-Zusammensetzung im Safran. Helv. Chim. Acta 58: 2233–2236.
Phillips, J.M. and Hayman, D.S. (1970) Improved procedures for cleaning roots
and staining parasitic and vesicular-arbuscular mycorrhizal fungi for rapid
assessment of infection. Trans. Br. Mycol. Soc. 55: 158–162.
Rosewarne, G.M., Barker, S.J. and Smith, S.E. (1997) Production of near-synchronous fungal colonization in tomato for developmental and molecular
analyses of mycorrhiza. Mycol. Res. 101: 966–970.
Schröder, S., Hildebrandt, U., Bothe, H. and Niehaus, K. (2001) Suppression of
an elicitor-induced oxidative burst reaction in Nicotiana tabacum and Medicago sativa cell cultures by corticrocin but not by mycorradicin. Mycorrhiza
11: 101–106.
Schwartz, S.H., Qin, X. and Zeevaart, A.D. (2001) Characterization of a novel
carotenoid cleavage dioxygenase from plants. J. Biol. Chem. 276: 25208–
25211.
Smith, S.E. and Read, D.J. (1997) Mycorrhizal Symbiosis. 2nd edn. San Diego:
Academic Press.
Vierheilig, H., Maier, W., Wyss, U., Samson, J., Strack, D. and Piché, Y. (2000)
Cyclohexenone derivative- and phosphate-levels in split-root systems and
their role in the systemic suppression of mycorrhization in precolonized barley plants. J. Plant Physiol. 157: 593–599.
Walter, M.H., Fester, T. and Strack, D. (2000) Arbuscular mycorrhizal fungi
induce the non-mevalonate methylerythritol phosphate pathway of isoprenoid biosynthesis correlated with accumulation of the ‘yellow pigment’ and
other apocarotenoids. Plant J. 21: 571–578.
(Received September 20, 2001; Accepted December 16, 2001)